Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    past auxiliaries and subjects

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     입니다
    1.10
    вают
    1.04
     있습니다
    1.02
    줍니다
    1.02
    큽
    1.01
    합니다
    1.01
    があります
    1.01
    입니다
    1.01
     требует
    1.01
    がありますが
    1.00
    POSITIVE LOGITS
     hadn
    2.37
     had
    2.30
     Had
    2.19
     había
    2.17
    had
    2.06
    Had
    2.02
     havde
    1.99
     hadde
    1.99
     habían
    1.94
     hade
    1.86
    Act Density 0.115%

    No Known Activations